Abstract:
Climate change has a crucial impact on livelihoods, economy, and water resources
due to the occurrence of weather and climate extreme events such as floods, droughts
and heat waves. Extreme weather has been increasing worldwide, hence the need to
understand their nature and trends. The aim of this study was to analyse the spatial
variability and long-term trends of climate extremes over the Limpopo in South Africa
from 1960 to 2014. Rainfall, temperature, and circulation fields were analysed to
understand the extent, nature of climate extremes over the Limpopo. Extreme value
theory (EVT) is a powerful method that was also employed in this study to provide
statistical models for events rarely observed. R statistical software was used for
clustering analysis which has a variety of functions for cluster analysis. Any station
whose value is larger than 95th for any day of the season was considered as a
widespread extreme event. The results show that the study area is highly vulnerable
to extreme events due to its latitudinal location and low altitude. Anomalous cut-off
lows, tropical cyclones and tropical storms are the major extreme producing systems
affecting the Limpopo province whilst the Botswana High becomes dominant during
heat waves and drought. Extreme weather events are common in Limpopo during
summertime and often coincide with mature phases of the El Nino Southern
Oscillation. In this study, after the suitable model for data was chosen, the interest was
in deriving return levels of extreme maximum rainfall. The computed data for return
levels predicted that the 5-year return period’s return level is approximately 223.89
mm, which suggests that rainfall of 223.89 mm or more per month should occur at that
station or location on the average of once every five years.
